Dataset Versions

v5

2023-10-06 3:11pm

Generated on Oct 6, 2023

Dataset Split

Train Set 83%
9736Images
Valid Set 8%
896Images
Test Set 10%
1132Images

Preprocessing

Auto-Orient: Applied
Resize: Fit (white edges) in 224x224
Auto-Adjust Contrast: Using Adaptive Equalization

Augmentations

Outputs per training example: 2
Flip: Horizontal
Shear: ±5° Horizontal, ±17° Vertical
Saturation: Between -26% and +26%
Brightness: Between -22% and +22%
Exposure: Between -18% and +18%
Noise: Up to 1% of pixels